Jobs to be Done – Accuracy in Real Time Location Systems


Jobs to be Done – Accuracy in Real Time Location Systems

While assisting in this white paper on how RTLS is empowering verticals, I had many realizations. There are several key ingredients to RTLS that are simply misunderstood.

RTLS is growing fast with an expected compounded annual growth rate (CAGR) of 48% until 2020. Still, we ignore the two key elements that actually make it valuable.

As RTLS and asset tracking use cases expand and penetrate more and more verticals, it is important to understand the features and trade-offs both users and buyers will need to consider.

To set the scene let’s consider two key elements: “time” and “location.”

In this post, we will focus on location and the actual real-world value of location information. In later discussion, we will explore important elements like tags, methods of deriving location information, data requirements, and the return-on-investment consideration. As all these decisions depend on the need for both time and location, we leave them to a later consideration.

Location & accuracy in RTLS

In most demos and presentations the location of an asset is indicated by a proverbial blue dot on a map. This dot is instantly recognizable in applications like Google Maps or in GPS systems. While this is intuitive and makes for impressive displays, it masks some of the important trade-offs that ultimately determine whether the application is successful and generates the expected returns.

Blue dot on a map

Succinctly put: A blue dot on a map makes sense whenever a human stands ready to interpret the information and take appropriate decisions.

While the information is sometimes useful, the number of relevant use cases is surprisingly small.

For example, when navigating with Google Maps, does the user care about where she is? Or is it more important to know she has to turn right in 150 meters to reach the destination in 30 minutes?

Even after experiencing a wide range of demos based on a variety of technologies and methods, one question is always raised: how accurate is it? This makes a lot of sense when looking at a demo. Plus, it is very a easy element to evaluate or verify. However, it ignores another, more important consideration: what does the user need that location information for?

When considering the relevance of accuracy, a good starting point is: who will use the information and what is to be achieved with it?

Zone or room level accuracy

Let’s imagine the popular use case of locating an IV pump in a hospital. What accuracy level gets the job done? Arguably, “the marginal return of location accuracy” drastically declines from room level downwards. In other words: once you know in which room the IV pump is located, it is probably easier to locate simply by looking around than enlisting technology. Human eyesight functions extremely well at room level.

Importantly, while the returns on accuracy decline after about 5 meters, the costs drastically increase.

Longitude and latitude positioning systems functioning at high accuracies usually require trilateration or triangulation which drives up the price of the required infrastructure and computing power. At the same time, it turns out that the real physical world in which we operate works mostly at a scale that comes down to room level accuracy or about 5 meter zones. Thus trade offs of accuracy and returns must be carefully weighed.

Location as an action trigger

So far we have assumed the user is a human consuming information from a blue dot on a map and then going into action. However, after a successful demo, it is much more likely that a computer will actually be the final user of the information.

For example, it is a computer that will trigger an alert when a person moves into a specific zone or room. A computer will record the transfer of a piece of equipment from one part of the facility to another.

A computer system is entirely ignorant about blue dots on maps. The information will most likely be consumed via an API. This fact frees us of any assumption that the worth of an RTLS technology directly correlates to the accuracy of its blue dot. The blue dot itself isn’t ROI positive. This also leads us toward to understand that the what and how of the location information usage is needed to create value.

Three questions for considering

When you consider implementing a Real Time Location System, ask yourself these questions before scheduling the first demo:

– What business metrics will generate returns on the system?

– What accuracy is required for the metrics that drive the business case?

– Who will interpret the location information?

Don’t forget: Accuracy is an expensive vanity metric unless it is necessary to get the job done.

There are several reasons to implement a real time location system. If you need help understanding the use cases, technologies, or, my personal favorite, the role of Bluetooth beacons, this white paper is both free and can help get you started.